College and Research Libraries DIANNE ELLSWORTH The Academic Library Looks at Union Lists Computer-based serials listings are rapidly becoming a popular method by which libraries make their serial holdings available. In addition, there is increasing pressure on libraries to participate in serial union lists. Because expenditure of staff time is an important factor in such participation, various kinds of union lists are examined in order to determine the most effective and least costly ways in which a library can have a list and attendant benefits for its own purpose, and also join in cooperative efforts. BY NOW, MANY LIBRARIES have either produced or are aware of the advan- tages of computer-based lists of serials for local needs. In addition, libraries realize that their participation will be requested in union list projects which are being planned on both regional and national levels. It is the purpose of this article to consider whether a library can satisfy the demands placed upon it for cooperative ventures and at the same time meet its individual needs. Three major developments have prompted such a discussion: ( 1) the proliferation of regional lists of serials; ( 2) the de- velopment of a program for a national serials data bank; and ( 3) the current state of computer t echnology as it re- lates to union lists. THE LOCAL LIST It is neither difficult nor uncommon for an individual library to produce a computerized list of serials. Most li- braries will feel that the time and ex- pense are worthwhile when compared Miss Ellsworth is serials librarian at W il- bur Cross Library, University of Connecti- cut, Storrs, Connecticut. with the limited products of a wholly manual operation. The most obvious ad- vantages of using a computer to per- form this operation are the ease of pro- ducing multiple copies and the possibili- ty of continuous updating. In addition, the data can be manipulated to produce lists by subject, vendor, or lists of in- complete files. 1 The serials lists thus be- come extremely useful for both the staff and clientele of the institution. Such a list can be distributed to other libraries who may wish to have it, but these individual lists are of limited use for the purpose of interlibrary loan. Producing a list at the local level allows the library considerable freedom in choosing the form of entry and type of programming best suited to its needs. THE uNION LIST Union lists of serials serve the dual purpose of providing bibliographic in- formation as well as identifying loca- tions where titles can be found. 2 If their growing number is any indication, union lists are much needed by libraries. It is useful at this point to examine the kinds of lists in which one can partici- pate, their possible cost, advantage, and problems for the large research library. I 475 476 I College & Research Libraries • November 1971 THE STATE LIST The state list serves the purpose of making known the resources of its pub- lic, special, and academic libraries. In such a list, therefore, similarity of col- lections is not the motivating factor. It has been pointed out that smaller li- braries in the state are more willing to cooperate in such a list since they will have less to report and will also benefit more because the burden of interlibrary loan is placed on the large library. 3 All large academic libraries in the state need to participate in order to ensure mutual benefits. Although both small and spe- cial libraries often have titles which the academic library might need, the library has to consider whether the cost of par- ticipation in terms of staff time matches the usefulness of the state list for the academic library. REGIONAL LISTS Regional lists can include any num- ber of libraries in various groupings. The groups can be arranged by similari- ty of collections and goals, by size of li- brary, etc. Regional lists can theoretical- ly provide a greater number of titles for interlibrary loan purposes than a state list. In addition, such a grouping can be the basis for cooperative acquisi- tions and storage and weeding programs. In order to be effective, however, the group must agree on common goals such as form of entry and the kinds and amount of titles which will be included. The costs for these activities should be shared so as to produce savings for all. SPECIAL LISTS An example of this kind is ACCESS- the list of scientific serials prepared under the jurisdiction of the American Chemical Society. A list of this type is of importance primarily to a defined group of people. The cost of participat- ing in such a list must be measured in terms of the need it fulfills for the in- dividual library. NATIONAL LISTS The Union List of Serials and New Serials Titles provide the widest general coverage of location and bibliograph- ical description now presently available. However, not all locations for a title are given, a factor which led in the past to the development of lists at the local and regional level. Currently in the develop- ment stage is the National Serials Data Program. The announcements for this project indicate that it will offer a wide variety of services, such as information on holdings and location for all serials, publishing and dissemination of special union lists by categories, and machine- readable data for local processing.4 While this program seems admirable in its definition and scope, it will probably be some time before it is fully opera- tional. The pilot project is dealing now with medical and scientific serials, al- though ultimately it will be expanded into other disciplines. 5 Operating within the national framework would provide certain advantages: ( 1) greater coverage of titles; ( 2) standardization (of for- mat); and ( 3) greater access to govern- ment programs and funds. COST OF pARTICIPATION The cost of participating in union list projects can be measured in the amount of staff time used in the report- ing of holdings and in the subsequent updating procedures. Although a library may be willing to cooperate in union list projects, it must seriously consider the economics of filling out numerous checking editions and continuously re- porting to diverse agencies. If there is any increasing demand for this kind of participation, a library would either have to hire additional personnel or di- Academic Library Looks at Union Lists I 477 vert present staff from the daily opera- tion which most libraries can ill-afford. Sending .a locally produced tape is usual- ly not feasible unless it is compatible with the master tape of the producer- agency. If the library's own list can be us ed as the basis of a union list, this is of cours e desirable, but cannot b e ex- pected. Given this situation, the follow- ing actions are possible: ( 1 ) the library can develop its own list and ignore other projects; ( 2) the library can develop its own list and participate in a few select- ed union lists which it feels it cannot ignore; ( 3) the library can set up guide- lines for participation in other projects and hopefully develop its own list in conjunction with one of these projects; and ( 4) the library can participate in all union list projects. There is no question but that there is some cost involved when a library de- velops a list on its own. Included is the staff time needed to prepare the records; the hiring or training of computer per- sonnel; the expense of running a com- puter; and supplies. If the library can obtain the information and services it needs in another way, this kind of indi- vidual expenditure would not be neces- sary. It is also true that .a library which has a large interlibrary loan operation will find it easier with a comprehensive union list at its disposal. The third al- ternative is, therefore, probably the most feasible. If a library has need of setting up guidelines for participation, one or more of the following criteria should influence the decision: ( 1) evenly bal- anced sharing of interlibrary loan re- quests; ( 2) the possibility of shared ac- quisition programs ; ( 3) the possibility of having lists produced as a substitute for local records (vendor lists, ceased titles); ( 4) the availability of lists of the individual library holdings by sub- ject to satisfy the demands of local users; (5) a format which can be uti- lized by different libraries (i.e., sufficient cross-references for variant forms of entry); and ( 6) representation in the decision-making process. The National Serials Data Progr am shows the most promise since it w ould lead to standardization, fulfill most of the criteria mentioned above, and great- ly reduce duplications effort. Most of this will be accomplished by utilizing computer technology. A recent article has pointed out, however, that advances in .automation have led thus far to an increase in regional lists rather than to processing at the national level. 6 If the national program is capable of provid- ing services for both large and small li- braries in the near future, it would seem wasteful to devote so much money .and effort to local solutions. However, the increasing d emands placed on libraries by their clientele and the need for interlibrary loan informa- tion have tended to favor the develop- ment of the regional plan. Many of the regional lists now b eing developed have b een planned so they can be adapted by other library systems. Such lists as the Union Catalog of Medical Periodicals (New York) and, in particular, the In- termountain Union List of Serials have no limitation to their capacity or flexi- bility.7 A proposal for a regional list in lieu of a state/ national/or local list at this time would have the following advan- tages: ( 1) utilization of work (pro- grams, tapes, etc.) done by other li- braries; ( 2) a central processing area, i.e., several libraries can absorb the cost of processing and editing the data; ( 3) use of a computer and computer per- sonnel primarily engaged in library r e- lated .activities-the cost shared by par- ticipating libraries; ( 4) a larger data base than could be provided by an in- dividual library or by the libraries with- in a single state; ( 5) interlibrary loan cooperation because of geographic prox- 478 I College & Research Libraries • November 1971 imity; and ( 6) possibility for coopera- tive acquisition plans to offset the rising cost of serials. In order for the center to be effective, the following requirements would have to be fulfilled: ( 1) defining a group of libraries which would be willing to co- operate and would benefit equally from the project; ( 2) utilization of programs compatible with the MARC/ COBOL format of the Library of Congress; ( 3) possibility of on-line interrogation to save time and transmit machine-read- able information; ( 4) ability of the program to add or suppress certain kinds of information to allow for local variation; ( 5) production of lists by subject categories; and ( 6) frequent up- dating. 8 A regional plan developed along these lines is NELINET, the joint venture of the five New England State University Libraries, which has already produced a shared cataloging program. The pro- gram utilizes the MARC II format and can provide individual services for par- ticipating libraries. A regional center can be established as a separate entity or can be located at a designated library, the participants sharing jointly in the cost of maintaining staff and equip- ment. A study of the costs and relative merits of the type of center would have to be made before any decision could be reached. In summary, the adoption of a re- gional plan at this time would include greater benefits for the individual li- brary by expanding the resources avail- able to it. Current computer technology can provide the flexibility needed to make the computer product useful for individual participants. Any proposal for a regional union list, whether on state or type-of-library level should con- sider the plans underway at the national level. In this way the regional center can act as a link in the network when the national program is operable by be- ing able both to accept from and com- municate information to the national center. If such steps as those listed above are taken, regional plans need not be looked at as an attempt to deflect en- ergy from efforts being made at the na- tional level, but as a positive step to- ward greater cooperation and commu- nication. At the same time, needed ser- vices will be provided to the faculty, students, and staff at the library. A library can only decide for itself to what extent it can cooperate with re- quests for union list participation, but a library which is involved in a working, cooperative project can afford to be more discriminating about its involve- ments. REFERENCES 1. Jacqueline W. Felter and Djoeng S. Tjoeng, "A Computer System for a Union Catalog: Themes and Variations," Medical Library Association BulleHn 53: 163- 69 ( April 1965 ). 2. Harry Dewey, "The National Union List of Serials: Weaknesses and a Proposal," Li- brary Resources & Technical Services 2: 225-38 ( Fall1968). 3. Kenneth D. Olson, "Union Lists and the Public Record of Serials," Special Libraries 61:15-20 (Jan. 1970). 4. William H. Huff, "The Acquisition of Serial Publications," Library Trends 18:314 (from a working paper prepared by Elaine W. Woods, National Serials Data Program, Phase I). 5. Association of Research Libraries, "Present Status of the National Serials Pilot Project" (Memorandum), 13 Jan. 1969, p.3. 6. Scott Adams, "Progress Toward a National Serials Data System," Library Trends 18: 512 (April 1970). 7. Felter, "A Computer System," p.177. 8. William R. Nugent, NELINET; the New England Library Information Network, Pa- per presented at the International Federa- tion for Information Processing, Edinburgh, 1968 (Cambridge, Mass.: Inforonics), un- paged.